Method for forming fine structure to volume of substrate made of brittle material

JP2023164878A5Inactive Publication Date: 2025-01-20SCHOTT AG
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
JP2023137349
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Priority Date
2018-04-27
Filing Date
2023-08-25
Publication Date
2025-01-20
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Existing methods for structuring glass and glass-ceramic substrates, such as laser ablation and sandblasting, result in microcracks, deformations, and reduced strength due to heat load and mechanical stress, limiting the ability to create fine, complex structures efficiently.

Method used

A method combining ultrashort pulse laser-induced filamentary damage and subsequent etching to form cavities in brittle materials like glass or glass-ceramics, where the laser creates non-penetrating filaments that are expanded by etching to form precise, complex geometries without penetrating the substrate.

Benefits of technology

This approach minimizes microcrack formation, maintains substrate strength, and allows for efficient production of fine, complex structures with reduced roughness and stress, enabling faster processing times and improved industrial applicability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 00000000_0000_ABST
    Figure 00000000_0000_ABST
Patent Text Reader

Abstract

To provide a method for forming a fine structure or notch on a glass element or a glass ceramic element.SOLUTION: A method is defined for forming a cavity in a substrate 1 made of a brittle material comprising glass or glass ceramics, in which a laser beam 27 of an ultrashort pulse laser 30 is directed onto one of the side surfaces 2, 3 of the substrate 1, and is focussed by a focussing optical element 23 to form an elongated focal point in the substrate 1, the incident energy of the laser beam 27 forms at least two filament-shaped lesions 32 in a volume of the substrate 1 to expose the substrate 1 to an etching medium, the etching medium removes a material of the substrate 1 at a removal rate of 2 μm / hour to 20 μm / hour and expands at least two filament-shaped lesions 32 to form filaments to bind at least two filaments to form one cavity.SELECTED DRAWING: Figure 1
Need to check novelty before this filing date? Find Prior Art

Description

[Technical Field]

[0001] The present invention relates to a method for forming fine structures or notches in a volume of a substrate made of generally brittle material, particularly in a glass element or glass-ceramic element. The present invention further relates to a glass element or glass-ceramic element having such a structure or notch, manufactured by this method.

[0002] Precise structuring of transparent, translucent, or opaque glass or glass ceramics is advantageous in many applications. In these cases, precision in the range of a few micrometers is required. Structuring relates to (round or square) holes, cavities, passages, or any freeform shapes. To enable use in a wide range of applications, it is desirable that the processing leaves no damage, residue, or stress in the edge region or volume of the substrate.

[0003] Furthermore, such structuring is desirable to enable the most efficient manufacturing process possible. Various methods can be used to create the holes. Besides sandblasting with a corresponding mask, ultrasonic machining is an established method. However, both methods are limited in terms of scaling to small structures, typically around 400 μm for ultrasonic machining and a minimum of 100 μm for sandblasting. Based on mechanical removal, sandblasting also introduces stress within the glass related to peeling in the hole edge region. Neither method is fundamentally suitable for structuring thin glass.

[0004] Therefore, recently, multiple laser sources are used for structuring various different materials. In this case, almost all known laser sources are used, such as CO2 lasers or CO lasers, and diode-pumped ns-, ps-, and fs solid-state lasers with wavelengths of infrared (e.g., 1064 nm), green (532 nm), and UV (365 nm). Excimer lasers working at extremely short wavelengths (e.g., 193 nm or 248 nm) are also used for processing.

[0005] U.S. Patent Application Publication No. 2015 / 0165563 (US2015 / 0165563A1) or U.S. Patent Application Publication No. 2015 / 0166395 (US2015 / 0166395A1) describes a laser method for forming through holes in thin glass or as preparation for subsequent separation of glass components.

[0006] The processing of glass or glass-ceramics is particularly in demand because glass or glass-ceramics generally have low thermal conductivity and high fracture sensitivity. Therefore, all laser ablation methods result in a more or less strong thermal load or heat introduction. This heat introduction can certainly be reduced by shorter wavelengths and shorter pulse lengths, but it still results in partially critical stresses leading to microcracks and deformation in the edge region of the hole. At the same time, this method also creates a clearly measurable roughness at the hole wall because all laser methods ablate in a clustered manner, meaning that the size of each cluster determines the residual roughness of the wall.

[0007] Furthermore, a drawback of the laser ablation method is that sufficiently deep structures can only be achieved by passing the workpiece through it multiple times. The processing speed is correspondingly slow. Therefore, this method is only conditionally suitable for use in industrial production. This is especially true when processing openings that penetrate glass, or structures that generally extend from one side to the opposite side. Moreover, the walls of such structures, such as grooves, are inclined, i.e., not vertical.

[0008] In particular, when structuring brittle materials such as glass or glass ceramics, another problem arises: the processed structure significantly reduces its strength under bending load. This is because microcracks are generated during structuring. Another drawback is the accumulation of removed material.

[0009] A method for processing a through-hole in a glass substrate is described in German Patent Application Publication No. 102013103370 (DE102013103370A1). In this method, an aperture is first formed in the glass substrate by a laser beam, and then material removal is performed by etching.

[0010] The drawback of this known method is that it is extremely difficult to manufacture fine and complex structures within the volume of a glass substrate or glass-ceramic substrate with acceptable effort and high quality (i.e., with respect to the strength of the components).

[0011] The inventor has dedicated himself to this problem. Therefore, the fundamental problem of the present invention is to provide a glass element or glass-ceramic element having at least one opening or notch, while minimizing the reduction in strength. In this case, it is desirable that the method for forming the opening in the glass element or glass-ceramic element results in as few microcracks as possible. At the same time, it is desirable that the process time of the method be improved.

[0012] In particular, it is desirable that the geometric shape for the notch can be freely selected. It is especially desirable that the notch formed in this form does not penetrate the substrate, that is, that the surface of the substrate on the opposite side of the notch remains unchanged.

[0013] This problem is unexpectedly easily solved by a method for forming a cavity in a substrate made of a brittle material, as described in any one of the independent claims, and by a glass substrate or glass-ceramic substrate having such a cavity. Preferred embodiments or improvements of the present invention can be found in the respective dependent claims.

[0014] Therefore, the present invention relates to a method for forming at least one cavity in the volume of a substrate made of a brittle material, preferably glass or glass ceramics.

[0015] When simply referred to as a substrate, it should be understood below as a substrate containing brittle materials. The substrate is advantageously made of or contains glass or glass ceramics.

[0016] The substrate according to the present invention may be formed in the shape of a plate having two sides located on opposite sides, for example, but this is not a prerequisite for carrying out the method. It is not essential that the substrate is formed to be particularly thin, because through-holes that limit the thickness of the substrate do not need to be formed. The substrate according to the present invention may have a thickness of 500 μm or more.

[0017] In the following, the concept of a "filament" is understood to mean an extremely fine blind hole, that is, an opening that does not penetrate the substrate. According to the present invention, a filamentous damage area is formed by the laser beam of an ultrashort pulse laser. Therefore, according to the present invention, a substrate having at least one filamentous damage area that does not penetrate the substrate can be obtained as an intermediate product. The diameter of this filamentous damage area is then expanded by an etching process, thereby forming a filament. Therefore, according to the present invention, a substrate having at least one more filament that does not penetrate the substrate can be obtained as an intermediate product.

[0018] The concept of "cavity" is hereinafter understood as an opening or notch that can be formed by the joining of at least two filaments adjacent to each other as a result of the extension of the filaments involved in the subsequent etching process. This opening faces towards one side of the substrate, that is, towards one of the side surfaces, and thus defines a notch on the surface of the substrate. The opening can take various different geometric shapes or complex structures. Therefore, the cavity is not a through-opening that extends from one side surface to the opposite side surface.

[0019] The method according to the present invention envisages the following steps: - Directing the laser beam of an ultrashort pulse laser towards one of the side surfaces of the substrate and focusing it by means of a focusing optical element to form an extended focus in the substrate, Forming a filament-shaped damaged part in the volume of the substrate by the incident energy of the laser beam, the filament-shaped damaged part entering into the volume until a predefined depth is reached, and in particular not penetrating the substrate, In order to form the filament-shaped damaged part, the ultrashort pulse laser emits a pulse packet having at least two laser pulses, either pulsed or continuous, After processing at least two filament-shaped damaged parts, - Exposing the substrate to an etching medium, the etching medium removing the material of the substrate at a removal rate of 2 μm per hour to 20 μm per hour, - Expanding at least two filament-shaped damaged parts to form filaments, - Joining at least two filaments, preferably at least 20 filaments, particularly preferably at least 50 filaments to each other to form one cavity.

[0020] According to the present invention, a substrate made of a brittle material, preferably a substrate including a glass element or a glass-ceramic element, further includes at least one cavity. This cavity, provided in the volume of the substrate, can be manufactured by a combined method including a laser-based method for forming a filamentous damage area and a subsequent etching method for expanding the damage area to form a cavity.

[0021] Preferably, at least two adjacent filamentous damage areas are fabricated in the substrate. Then, the etching process can remove the material between at least two filamentous damage areas together, thereby forming a cavity. Since the material is removed by the etching process, the dimensions of the cavity are larger than the filamentous damage areas or filaments. Therefore, preferably, the cavity includes at least two filamentous damage areas, thereby forming a notch.

[0022] This was unexpected because conventional etching methods, that is, the expansion of extremely small filamentary damage areas in the material by an etching medium, had been assumed to require through-holes or openings with a relatively large diameter. Indeed, such through-holes also require surface damage to the opposite side of the substrate. Furthermore, there are limitations regarding the thickness of the substrate, which are not applicable to this invention.

[0023] The requirement for through-holes is based on the assumption that the etching medium can only enter such small filamentous damage areas, since it is drawn into the filamentous damage area from the first side by capillary force, and in this case, gas present in the filamentous damage area can leak out on the other side located opposite.

[0024] Unexpectedly, however, the inventors discovered that the etching solution can penetrate into filamentous damaged areas, that is, into openings or blind holes that do not penetrate the substrate, and expand these openings or blind holes. Moreover, according to the present invention, it is possible to expand filamentous damaged areas that do not penetrate the substrate, having a diameter of up to 1 μm, preferably up to 0.8 μm, and particularly preferably up to 0.5 μm, measured in the substrate volume, using the etching medium. Therefore, when the substrate or a region of the substrate provided for expansion is exposed to the etching medium, the opening to be expanded does not penetrate and extend from one side to the other. Consequently, the surface located on the side opposite to the laser beam entry side remains unchanged.

[0025] If filamentous damage areas formed by the laser method are located close together, subsequent etching methods can dissolve the remaining wall material between two adjacent filamentous damage areas, potentially forming a complex cavity.

[0026] The pulse energy of individual pulses is selected so that this pulse energy is below the ablation threshold of the substrate material, allowing the laser light to penetrate the material and preventing the laser energy from being consumed by the ablation process already occurring on the surface.

[0027] Therefore, when filamentous damage areas are formed in a series, a cavity can be formed in the substrate volume particularly well by processing the filamentous damage areas and subsequent etching. The cavity can be formed with various different taper angles. In this case, the taper angle is the angle of the cavity wall relative to the adjacent side surface.

[0028] The taper angle is determined in particular by the removal rate in the etching process. At high removal rates, such as in the range of 15 μm to 20 μm per hour, a somewhat vertical configuration of the cavity remains, resulting in a taper angle in the range of 90° ± 5°, preferably 90° ± 3°, and particularly preferably 90° ± 1°. Significant deviation from the vertical configuration of the cavity occurs with slower etching processes. This relates to removal rates in the range of 2 μm to 10 μm per hour.

[0029] The depth and extent of the filamentous damage in the substrate volume caused by the incident energy of the laser beam can be controlled by intentionally adjusting the focal position relative to the substrate. For this purpose, a corresponding optical element may be provided, which can control the focal position and depth.

[0030] To achieve an extended, uniform damage area to a predetermined depth within the substrate volume, a burst mode with pulse packet incidence is particularly advantageous. This specifically means that the filamentous damage area does not penetrate the entire substrate, but extends only to a predetermined portion of the substrate thickness at the aforementioned location. It is specified that the filamentous damage area penetrates less than 80%, preferably less than 70%, and particularly preferably less than 50% of the substrate thickness. In other words, the length of the filamentous damage area is less than 80%, preferably less than 70%, and particularly preferably less than 50% of the substrate thickness at the location of the damage.

[0031] Furthermore, it should be considered that the subsequent etching process will also remove other material in the bottom region of the filamentous damage. In this case, it is considered advantageous if the filamentous damage penetrates the substrate to a depth of less than 80%, preferably less than 70%, with respect to the substrate thickness, in order to ensure that preliminary damage occurs on the opposite side.

[0032] As the etching medium, an etching solution comparable to that used in plasma etching is preferred. That is, according to this embodiment, etching is carried out wet chemically. This is advantageous for removing substrate components from the surface during etching. Acidic and alkaline solutions can be used as etching solutions. Suitable acidic etching media include HF, HCl, H2SO4, ammonium hydrogen fluoride, HNO3 solutions, or mixtures of these acids. For basic etching media, KOH- or NaOH alkaline solutions are preferably considered. Acidic etching solutions typically achieve high removal rates. Basic solutions are preferred, in particular, when slow removal is desired. Furthermore, etching is preferably carried out in a temperature range of 40°C to 150°C, preferably 50°C to 120°C, and especially preferably up to 100°C.

[0033] In general, silica-containing glass with low alkalinity is particularly suitable for structuring according to the present invention. Too high an alkalinity makes etching difficult. It has been found to be advantageous to use glass with a basicity in the range of 0.45 to 0.55, preferably 0.48 to 0.54. This makes the glass suitable for controlled etching, especially with basic etching media. However, etching with acidic etching media remains possible. According to another improved embodiment of the present invention, the glass of the glass element is specified to be silica glass containing less than 17 weight percent of alkali oxide.

[0034] However, this method also makes it possible to form cavities in glass ceramics, particularly using the LAS system.

[0035] In the burst operation mode defined for this invention, laser energy is emitted not as individual pulses, but as a series of short pulses emitted in succession, which together form a single pulse packet, a so-called burst. Such pulse packets typically have somewhat greater energy than individual pulses in normal single-shot operation. However, the pulses of the burst themselves contain significantly less energy than individual pulses. With respect to pulses in a burst, the pulse energy may be flexibly adjustable, and in particular, it may be defined that the pulse energy remains substantially constant, or that the pulse energy increases or decreases.

[0036] A suitable laser source according to the present invention is a neodime-doped yttrium-aluminum-garnet laser having a wavelength of 1064 nanometers. Operation with second harmonic generation (SHG) or third harmonic generation (THG) is also possible. When selecting a laser source, it should be noted that the substrate material is at least partially transparent to the corresponding wavelength of the laser beam.

[0037] The laser source is, for example, 12 mm (1 / e 2 A raw beam with a diameter of ) can be generated, and a biconvex lens with a focal length of 16 mm can be used as the optical element. For generating the raw beam, an optical element that appropriately forms the beam, such as a Galilean telescope, can be used in some cases. The laser source operates at repetition rates between 1 kHz and 1000 kHz, preferably between 2 kHz and 200 kHz, and especially preferably between 3 kHz and 100 kHz.

[0038] The repetition rate and / or method speed, i.e., the speed at which the laser beam is guided relative to the substrate, can be selected so as to achieve a desired spacing between adjacent filamentous damaged areas.

[0039] The appropriate pulse width of the laser pulse is in the range of less than 100 picoseconds, preferably less than 20 picoseconds. The typical power output of the laser source is particularly advantageous in the range of 20 to 300 watts. To obtain a filamentous damage area, according to an advantageous modification of the present invention, a burst pulse energy higher than 400 microjoules is used, and more advantageously, a total burst energy higher than 500 microjoules is used.

[0040] When operating an ultrashort pulse laser in burst mode, the repetition rate is the repetition rate of burst emission. The pulse width is substantially independent of whether the laser is operated in individual pulse mode or burst mode. Pulses in a burst typically have a pulse length similar to that of pulses in individual pulse mode. The burst frequency is in the range of 15 MHz to 90 MHz, preferably in the range of 20 MHz to 85 MHz, for example, 50 MHz, and the number of pulses in a burst is 2 to 20 pulses, preferably 3 to 8 pulses, for example, 6 pulses.

[0041] The filamentous damage may consist of a series of localized defects arranged in sequence, for example, a row of filamentous damage that reaches deep into the substrate. This row of filamentous damage penetrates the substrate to a predetermined depth. Therefore, it is not necessary to use a particularly thin substrate.

[0042] The present invention will be described in detail below with reference to the accompanying drawings, with regard to preferred embodiments. [Brief explanation of the drawing]

[0043] [Figure 1] This figure shows a device for laser processing glass elements in preparation for subsequent etching. [Figure 2] This is a top view of a glass element with a processed, filamentous damaged area. [Figure 3a] This is a magnified image of an electron microscope photograph of the edge of a glass element. [Figure 3b] This figure illustrates an electron microscope image of the cavity wall. [Figure 4] This is a side view showing the arrangement of filamentous damaged areas within the volume of the substrate. [Figure 5] This is a top-down plan view of the side of a substrate having a matrix-like arrangement of filamentous damaged areas. [Figure 6a] This diagram shows the mutual influence of already formed adjacent filamentous damage areas on the formation of new filamentous damage areas. [Figure 6b] This diagram shows the mutual influence of multiple already formed, adjacent filamentous damage areas during the formation of new filamentous damage areas. [Figure 7a] This diagram shows the material-dependent relationship for two chains of filamentous damage that extend close to each other along two straight lines. [Figure 7b] This figure shows an arrangement for measuring the shielding effect of a chain of multiple filamentous damaged areas along a single straight line. [Figure 8a] This figure shows a three-step purification method for forming filamentous damaged areas that are spaced very close together. [Figure 8b] This figure shows the relationship between the distance between two filamentous damaged areas and the expected filament depth. [Figure 9a] This diagram shows the processing of the filamentous damaged area on the back side to form a back cavity. [Figure 9b] This diagram shows the processing of the filamentous damaged area on the back side to form a back cavity. [Figure 10a] This figure shows an embodiment for forming cavities in a substrate at different depths. [Figure 10b] This figure shows an embodiment for forming cavities in a substrate at different depths. [Figure 11a] This is a perspective view of each of the substrates having the surface structure formed by the present invention. [Figure 11b]This is a perspective view of each of the substrates having the surface structure formed by the present invention. [Figure 11c] This is a perspective view of each of the substrates having the surface structure formed by the present invention. [Figure 12] This is a top-down plan view of a portion of a glass element that has a cavity. [Figure 13a] This is a schematic diagram showing another assembly of two glass wafers for use as a microfluidic chip. [Figure 13b] This is a schematic diagram showing another assembly of two glass wafers for use as a microfluidic chip. [Figure 14] This figure shows an example of a microfluidic cell. [Figure 15] This figure shows an example of a microfluidic cell. [Figure 16] This figure shows an assembly for applying an adhesive layer.

[0044] Detailed description of preferred embodiments In the following detailed description of preferred embodiments, for clarity, the same reference numerals indicate substantially identical parts in those embodiments. However, for clarity, the preferred embodiments illustrated in the drawings are not necessarily shown to scale.

[0045] Figure 1 shows an embodiment for a laser processing apparatus 20. This laser processing apparatus 20 can process a substrate, such as a glass element 1, into a filamentous damage portion 32. This then allows for the formation of a cavity from the filamentous damage portion 32 in an etching process.

[0046] The apparatus 20 includes an ultrashort pulse laser 30 equipped with a pre-positioned focusing optical element 23 and a positioning unit 17. This positioning unit 17 allows the collision point 73 of the laser beam 27 of the ultrashort pulse laser 30 to be positioned laterally on, for example, the side surface 2 of a plate-shaped glass element 1 to be processed. In the illustrated example, the positioning unit 17 includes an xy table. One side surface 3 of the glass element 1 is placed on this xy table. However, alternatively or additionally, the optical element can be configured to move in order to move the laser beam 27, thereby allowing the collision point 73 of the laser beam 27 to move when the glass element 1 is fixed.

[0047] The focusing optical element 23 focuses the laser beam 27 in a beam direction perpendicular to the side surface 2 to be illuminated, thereby forming an extended focal point. Such a focal point can be formed, for example, by a conical lens (so-called axicon) or a lens with large spherical aberration. The positioning unit 17 and the ultrashort pulse laser 30 are preferably controlled by a programmatically controlled computing unit 15. This configuration allows for reaching a predetermined position for processing the filamentous damage portion 32, which is preferably achieved by reading position data from data sources or via a network.

[0048] In the illustrated example, the illustrated filamentous damage portion 32 penetrates to a predetermined depth in the volume of the glass element 1. In this example, this depth corresponds to approximately half the thickness of the plate-shaped glass element 1. According to the present invention, the length of the filamentous damage portion 32 is less than 80% of the thickness of the substrate at the location of the processed filamentous damage portion 32, preferably less than 70%, and particularly preferably less than 50%.

[0049] According to one embodiment, the following parameters can be used for the laser beam.

[0050] The wavelength of the laser beam is typically 1064 nm for an Nd:YAG laser. A laser beam with a raw beam diameter of 12 mm is formed, and this laser beam is then focused by an optical element in the form of a biconvex lens with a focal length of 16 mm. The pulse width of the ultrashort pulse laser is less than 20 ps, ​​preferably about 10 ps. The pulses are emitted in bursts of 2 or more, preferably 4 or more pulses. The burst frequency is 12 to 48 ns, preferably about 20 ns, the pulse energy is at least 200 microjoules, and the burst energy is correspondingly at least 400 microjoules.

[0051] Next, after processing one or particularly many filamentous damage areas 32, the glass element 1 is removed and placed in an etching bath. In this etching bath, the substrate material, i.e., glass in this embodiment, is removed along the filamentous damage areas 32 in a slow etching process, so that filaments are formed within the volume of the glass element 1 at the locations of such damage areas 32.

[0052] Preferably, a basic etching bath having a pH value of >12, such as a KOH solution with a pH of >4 mol / l, preferably >5 mol / l, particularly preferably 6 mol / l, but <30 mol / l, is used. Etching is carried out at an etching bath temperature of >70°C, preferably >80°C, particularly preferably >90°C, according to one embodiment of the present invention, regardless of the etching medium used.

[0053] Figure 2 shows a glass element 1 having numerous filamentous damages 32, in a plan view of the side 2 seen from above. The filamentous damages 32 are arranged in a defined pattern so that they can be written into the glass element 1 by the computer-controlled positioning unit 17 and the ultrashort pulse laser 30 as described above. In particular, the filamentous damages 32 are machined into the glass element 1 in this case, for example, along a closed rectangular contour or a defined path of shape. The corners of the lines may be slightly chamfered. It will be obvious to those skilled in the art that this method can be used to follow not only rectangular paths but also arbitrarily formed paths.

[0054] In the embodiment that forms the basis of the method according to the present invention, - The collision point 73 of the laser beam 27 on the substrate is guided along a defined path or to a predetermined position. -By intentionally forming a focal position relative to the substrate at a defined position, a filamentous damaged portion 32 is formed to a defined depth in the volume of the substrate. Next, etching expands the filamentous damaged area 32, forming a filament within the volume of the substrate.

[0055] In a particular embodiment, the optical element 23 is configured to be similarly controllable, so that the precise focal position and depth in the substrate can be controlled particularly easily. This configuration makes it possible to form the filamentous damage portion 32 at various heights in the substrate.

[0056] The diameter of the filamentous damaged area 32 is expanded by subsequent isotropic etching. In this case, the cavity can be formed by connecting at least two adjacent filaments in the subsequent etching process. The steep angle, or taper angle, of the filamentous damaged area 32 is maintained during slow isotropic etching, i.e., at a low removal rate.

[0057] According to the present invention, this extension for integrating adjacent filaments is desired to form a cavity 5 having a complex geometric shape and / or to form a cavity 5 containing multiple filaments.

[0058] Figure 3a shows an electron microscope image of a longitudinal section of a series of filamentous damage areas 32 arranged in a row on a glass element 1 processed according to the present invention. An edge can be seen having numerous filaments 6 that extend parallel to each other and open to the side. These filaments 6 are, for example, elongated in shape. In the illustrated example, the filaments 6 are formed by an etching process at the location of the former filamentous damage area. Integration of adjacent filamentous damage areas has not yet been performed in this example. In the drawing shown in Figure 3, in a plan view of the cut edge from above, it can be seen that the longitudinal direction 51 of the filaments 6 extends into the volume of the glass element 1, particularly in the vertical direction, starting from one side surface 3.

[0059] Figure 3 further allows for a clear view of the microstructure of the filament 6, specifically the shape of the sidewalls of the filament 6, which are in the form of spherical or chamfered cap-shaped recesses 7. Preferably, through a slow etching process, these spherical recesses 7 are adjacent to each other, and the abutting concave rounded portions of the recesses 7 form ridges.

[0060] In this example, the spacing between the filaments 6 is relatively large, approximately 50 μm. This spacing can also be selected to be smaller, especially if the filaments 6 transition directly to each other without flat edge divisions 11. Generally, the spacing between the filaments 6 (also called the "pitch") is preferably in the range of 3 μm to 70 μm, preferably at least 10 μm, and particularly preferably at least 20 μm. This spacing is measured from center to center of the filament 6.

[0061] Figure 3b shows an illustrative electron microscope image of the wall portion 8 of cavity 5. The spherically missing surface of the wall portion 8 can be clearly seen. The illustrated example is the wall portion 8 of cavity 5 processed according to the present invention from transparent, colorless borosilicate glass, such as D263T from Schott AG (located in Mainz). The expansion of the pre-processed filamentous damage portion 32 was carried out at an etching medium temperature of 80°C for 19 hours.

[0062] Figure 4, which schematically shows a longitudinal cross-sectional view of the glass element 1 that has already been processed and has a total of five filamentous damage portions 32 in this example, clearly shows that the filamentous damage portions 32 do not completely penetrate the glass element 1, but merely form blind holes.

[0063] According to the present invention, an etching medium is applied to the filamentous damaged area 32. This is unique in that the etching solution can penetrate into the filamentous damaged area 32, even though it is formed as a blind hole, and cause uniform, i.e., isotropic expansion. The filamentous damaged area 32 has extremely small dimensions. The filamentous damaged area 32 forming the basis of the wall portion 8 shown in Figure 3b of the cavity 5 has a maximum diameter of only 1 μm. Even smaller diameters, preferably up to 0.8 μm, and particularly preferably up to 0.5 μm, do not hinder the etching process according to the present invention.

[0064] In a particularly preferred embodiment of the present invention, a plurality of filamentous damaged portions 32 are processed into the glass element 1 in a series. The length t of the filamentous damaged portion 32 in the glass element 1, the size of the filamentous damaged portion 32 in cross-section, and the distance d between two adjacent filamentous damaged portions 32 are... x Depending on the configuration, this method allows for the formation of cavities 5 with complex geometric shapes, i.e., notches of various basic shapes or cross-sections, within the volume of the substrate.

[0065] FIG. 4 illustrates some important geometric relationships related to the configuration of the filamentous damage portion 32 within the glass element 1. In this example, five filamentous damage portions 32 of length t are shown arranged in parallel rows within the volume of a substrate having a thickness T, such as the glass element 1.

[0066] [[ID=__4]]As described above, since the filamentous damage portion 32 does not completely penetrate the substrate, t < T. To accelerate the processing of the filamentous damage portions into the substrate, the length of the filamentous damage portions is preferably selected to be as small as possible. This is because the processing of relatively long filamentous damage portions is extremely time-consuming in order to form a complex structure that requires such a large number of filamentous damage portions. Thus, preferably t < 0.8 * T, preferably t < 0.7 * T, particularly preferably t < 0.5 * T.

[0067] In addition to the depth in the substrate, the spacing between two adjacent filamentous damage portions is extremely important. This spacing is measured from the center to the center of the filamentous damage portion 32 and is shown as d in FIG. 4. x Thus, this "pitch" forms a reference for process control.

[0068] To form a three-dimensional inner structure within the volume of the substrate, characterized in that it has only one opening facing one of the two side surfaces of the substrate and the surface of the side surface located on the opposite side of the substrate remains unchanged, the filamentous damage portions 32 are processed within the volume of the substrate in a two-dimensional structure or pattern starting from one side surface. [[ID=__22]]

[0069] [[ID=__23]] FIG. 5 schematically shows a plan view from above of a side surface of a substrate having a matrix-like arrangement of filamentous damage portions 32, which in this example is a 4x5 arrangement. Due to this geometric arrangement of the filamentous damage portions 32, the inner contour that can be formed after the etching process is rectangular in shape as long as the spacing and depth of the filamentous damage portions 32 meet the above requirements. It will be apparent to those skilled in the art that various different geometric shapes can be formed based only on the arrangement of the filamentous damage portions 32 on the side surface of the substrate, for example, an inner contour within a glass element 1 having a triangular or circular basic shape or a free-form surface can be formed.

[0070] In FIG. 5, the two-dimensional spacing between two adjacent filamentous damage portions 32 is described by d x and d x perpendicular to d y and. In the matrix-like arrangement of the filamentous damage portions 32, these spacings may be the same in both directions, but this is not essential, so the condition of d x ≠d y is also possible. These spacings between adjacent filamentous damage portions 32 result in the defined formation of the bottom and wall portions 8 of the cavity 5 during the etching process. This cavity 5 is further deepened and advanced downward. In the example shown, d x =d y is selected.

[0071] In order to form a cavity having a desired structure by expanding the filamentous damage portions 32 by an etching process, it is important that the filamentous damage portions 32 are completely formed during the processing of the substrate material.

[0072] The formation of the filamentous damage portion 32 can be undesirably affected by contamination on the substrate surface. This can cause laser beam shading, which undesirably affects the beam shape or beam intensity. As a result, undesirable deviations occur in the formation of the filamentous damage portion 32 on the substrate, meaning that the specified depth and / or predetermined orientation cannot be achieved.

[0073] While it is true that a small gap between two adjacent filamentous damaged areas 32 is considered advantageous in forming a relatively large structure and minimizing the time required for material removal between the two adjacent filamentous damaged areas 32, care must be taken to maintain a specified minimum gap so as not to allow the filamentous damaged areas 32 to be too spatially close to each other.

[0074] This is because the spatial proximity of adjacent filamentous damage areas 32 to the current working point, i.e., the exact location where the new filamentous damage area 32 is being formed, has a significant influence on the shaping of the filamentous damage area 32 being formed, based on the shielding effect of the adjacent filaments. In optical structures working with spherical aberration or Bessel beams, the formation of the filamentous damage area 32 depends on the supply of laser energy, which is as rotationally symmetrical as possible with respect to the focal line.

[0075] For example, when a first chain is formed along a path by a continuity of filaments, the energy supply for forming a new filamentous damage 32a is hindered only by the already formed adjacent filamentous damage 32b. This is illustrated purely illustratively in Figure 6a, which aims to clearly show the mutual influence of already formed adjacent filamentous damages on the formation of a new filamentous damage.

[0076] The influence of adjacent filamentous damage areas 32b on the formation of a new filamentous damage area 32a depends overall on the spacing between the filamentous damage areas 32, 32a, and 32b, the diameter of the formed filamentous damage area 32, and consequently, the solid angle ratio that the existing filamentous damage areas 32b shading the energy supply to the new filamentous damage area 32a. In this example, this range is illustrated by angle δ. Each influence zone 33 of the laser beam is shown by a circle in this example. This circle shows the outer definition starting from the midpoint corresponding to the collision point 73 of the laser beam 27. Reference numeral 34 indicates an influence region that is shaded because it at least partially overlaps the influence zone 33 of the previously processed filamentous damage area.

[0077] Another important size is the diameter of the laser beam forming the filament on the sides 2,3 of the substrate. If a new filamentous damage area 32a is formed adjacent to multiple already formed filamentous damage areas 32b, for example, next to an already existing row of filamentous damage areas 32b, then if they are too close together, theoretically half of the energy supplied will already be shielded in several stereometric angles.

[0078] This is illustrated purely illustrative in Figure 6b, which illustrates the mutual influence of multiple already formed adjacent filamentous damage areas 32b during the formation of a new filamentous damage area 32a. In this case, the shading of individual solid angle ratios δ is added together to form overall shading. Therefore, overall, the maximum achievable length of a filamentous damage area 32a in the vicinity of adjacent filamentous damage areas 32b is a function of the distance between the filamentous damage areas. Knowledge of this functional relationship is used, according to the present invention, for optimizing the process strategy and, consequently, for controlling the method for forming cavities according to the present invention.

[0079] This relationship is material-dependent and will be briefly outlined in Figures 7a and 7b. In order to determine the material-dependent relationship, the first step is to identify the shielding effect of individual filamentous damaged parts 32a (Figure 7a). To do this, first along the first straight line 35, at a constant interval d from each other x A chain having filamentous damaged portions 32a with radius R that are spaced apart is processed onto the substrate. x >>R. Preferably, the spacing between the filamentous damaged portions 32a is selected to be large enough so that they are not affected by the already existing filamentous damaged portions 32a. This precondition is d x >50 * R, preferably d x >100 * This is given for the case where R is true.

[0080] Along the second straight line 36, a second chain of filamentous damaged portions 32b is formed at the same x position as the first chain, but with a constantly decreasing interval d. y It has the following starting value: y =d x A second chain of filamentous damage is formed, in which case the second straight line 36 to which it belongs extends continuously toward the first straight line 35. The optimal spacing can be determined by measuring the length of the formed filamentous damage in relation to the inter-spacing of the filamentous damage and may be used for another method. The length can be determined, for example, by optical microscopy, for example by opening the filament line, or by photography related to image processing, for example by edge extraction algorithms of edges.

[0081] Therefore, the optimal spacing between two adjacent filamentous damaged areas 32a and 32b is the minimum spacing d at which a reduction in the length of the filamentous damaged areas due to mutual influence has not yet occurred, or at which no difference in etching behavior has yet been observed. y That is the case.

[0082] In a similar manner, the shielding effect of a chain consisting of numerous filamentous damaged sections 32 along a straight line 35 (Figure 7b) can be measured. As in the example above, first, along the straight line 35, at regular intervals d x A chain of filamentous damaged portions 32 with a radius R, spaced apart, is processed onto the substrate. Then, various different intervals d y Individual filamentous damaged portions 32 are formed, and the length of each filamentous damaged portion 32 is newly determined depending on its distance from the straight line 35.

[0083] In certain cases, it may be necessary to form the filamentous damaged portions 32 at intervals smaller than those found to be advantageous based on the shielding measurements described above. This means that an interval is selected that may result in a reduction in the length of the filamentous damaged portions.

[0084] For this purpose, the methods illustrated in Figures 8a and 8b can be used particularly advantageously. For example, a method comprising multiple steps can be selected to produce a chain of filamentous damaged portions 132 along the path at a closer interval to one another in this context.

[0085] In the first step, a chain of filamentous damaged portions 132c is formed along a roughly defined path for the maximum possible length and at the minimum spacing between them. In this case, the minimum spacing is the distance between two adjacent filamentous damaged portions 132 that does not yet cause mutual influence when irradiated with a laser beam.

[0086] In the second step, another new, smaller filamentous damage portion 132b is formed in the center between each of two adjacent filamentous damage portions 132c. This cannot be done at the original length due to shielding by the adjacent filamentous damage portions 132c. If necessary, this process is recursively purified.

[0087] Figure 8a illustrates this method, which has three purification steps. Along the path, in this example, a straight line of length b, a filamentous damaged section 132c with a maximum depth t1 is first formed. This filamentous damaged section 132c arises based on a maintained minimum interval d1. In other words, step A involves the formation of the longest filamentous damaged section 132c.

[0088] Another new filamentous damage section 132b with a depth t2 is formed precisely in the middle of each of the two adjacent filamentous damage sections 132c. This filamentous damage section 132b has a spacing d1 / 2 = d2 on both sides relative to the adjacent filamentous damage sections 132c. In other words, this second step B involves the formation of a filamentous damage section 132b that is shorter than the filamentous damage sections 132c.

[0089] Finally, in the third step C, another new filamentous damage section 132a with a depth t3 is formed between each of two adjacent filamentous damage sections 132b. This filamentous damage section 132a has a spacing of d2 / 2=d3 on both sides relative to the adjacent filamentous damage sections 132b. This configuration allows for particularly narrow continuity of filamentous damage sections 132. In this case, each filamentous damage section 132 has a depth corresponding to each step of its formation. Figure 8b shows the relationship between the spacing d of two filamentous damage sections 132 and the expected filament depth t=t(d). In this case, the two adjacent filamentous damage sections 132 have different depths from each other.

[0090] As shown in Figure 8a, a cavity 5 with width b and a maximum depth t1, including the size expanded by the etching process, can be formed after the etching process is completed.

[0091] Naturally, the method can be carried out with more iterations than the three iterations described above, thereby achieving even narrower spacing between the filamentous damage areas 132. In this case, multiple filamentous damage areas 32,132 are fabricated on the substrate at the minimum spacing before shielding is performed, and then in one or more iterations, another filamentous damage area is formed in the center between two adjacent filamentous damage areas. In this case, the depths of the adjacently arranged filamentous damage areas are not generally the same and depend on each iteration in the method process in which the filamentous damage areas were manufactured.

[0092] In principle, the length of the focal line formed by the beam-forming optical element may exceed the thickness of the substrate. However, in this case, care should be taken to ensure that the side of the substrate opposite the entrance is not altered, because otherwise, undesirable damage may occur on the back surface of the substrate during the subsequent etching process. This can be countered, for example, by changing the distance between the optical element 23 of the focusing system and the substrate, that is, by increasing the distance of the optical element 23 from the surface of the substrate or, in this example, the glass element 1, and / or by appropriate refocusing by the optical system.

[0093] This is further influenced by the spacing of the filamentous damage. The spacing further depends on the specified depth of the cavity and the optical system used. Optical systems with spherical aberration, for example, an optical element in the form of a biconvex lens with a focal length of 16 mm, a free aperture of 18 mm and 12 mm (1 / e) at 1064 nm. 2 When using a raw beam and a desired cavity depth of up to 0.5 mm, it was found to be advantageous for the filamentous damaged portions 32 to be spaced apart from each other to be at least 10 μm, preferably 20 μm.

[0094] It is also possible to form a cavity 5 with a relatively large depth of up to approximately 5 mm. The depth of cavity 5 may be selected to be smaller. In this case, a minimum depth of at least 50 μm, preferably at least 100 μm, and particularly preferably at least 200 μm is advantageous so that the advantages of the method according to the present invention can be fully utilized for ablation as well.

[0095] The spacing of the filamentous damage also affects the shaping of the cavity bottom. In this case, two basic embodiments can be distinguished. 1) The filamentous damaged areas are processed into a matrix-like mesh (Raster) with equal spacing between them. In this case, d x =d y The etching process forms the bottom of the cavity, which has a uniform corrugated structure. This corrugated structure is formed between substantially selected intervals. x or d y It depends on. 2) The filamentous damage is processed in a continuous line. In this case, the filamentous damage is processed along adjacent paths with a first spacing of approximately 4-6 μm between them in the first direction. Another path of filamentous damage is added to the first path with a second spacing. In this case, the spacing between the paths is selected to be greater than the first spacing. For example, this second spacing may be 10 μm or more. The cavity formed in this manner has a bottom with a groove structure after the etching process.

[0096] Naturally, in the embodiments described above, the arrangement of the paths is selected according to the desired shape of the cavity. That is, the arrangement may be in any pattern, for example, a matrix shape. The chains or paths used to form the cavity do not need to be straight in this case, but can be selected in any free form.

[0097] In both embodiments 1) and 2) described above, the structure formed at the bottom of the cavity is essentially generated by the interaction between material alteration by an ultrashort pulse laser and the etching process. This bottom structure is characteristic of the filament formation strategy.

[0098] Both embodiments 1) and 2) mean that the wall portion 8 of the cavity 5 has a structure that includes a number of spherical recesses 7 after the etching process, as shown in Figure 3b, for example.

[0099] In a particular embodiment, the cavity 5 formed in the volume of the substrate according to the present invention includes at least two regions having different depths. In this embodiment, filamentous damaged portions 32, 132 are fabricated in the substrate at different lengths and / or different focal positions.

[0100] In the so-called surface cavity described above, where cavity 5 is formed on the substrate on the laser incidence side, it was found to be advantageous to first process relatively deep and long filamentous damage areas 32,132, and then relatively short filamentous damage areas. When forming the so-called back cavity, which is formed on the side of the substrate opposite to the laser incidence side, it was found that the reverse order is advantageous.

[0101] In general, the formation of a back cavity was found to be advantageous because a smaller filament spacing can be selected. Since less contamination occurs on the incident side due to the laser process, the filaments can be processed better at a specified depth and direction. Figures 9a and 9b show the processing of the back-side filamentous damage 32 to form a back cavity. In this case, a cavity 5 with a bottom is formed. In this case, the bottom has a groove-like shape in the embodiment shown in Figure 9a and a uniform wavy shape in the embodiment shown in Figure 9b.

[0102] In principle, as shown in Figure 10a, it is possible to form multiple cavities with different depths in the substrate, or even to form stepped sections. These stepped sections may be used in combination with through-holes 37 as suggested in Figure 10b to form blind holes, that is, the stepped sections have a cross-sectional shape with concentric cross-sections.

[0103] Unlike those shown in Figures 10a and 10b, preferably the spacing between filamentous damaged areas is adjusted so that the filaments are formed to their full length. In other words, the minimum spacing between two adjacent filamentous damaged areas is maintained without affecting the formation of each.

[0104] In a similarly advantageous embodiment, when the filamentous damaged portion is processed, at least one interface of the substrate is moistened with a liquid. In this case, various different embodiments are used. 1) Wetting the back surface of the substrate, i.e., the side opposite to the laser incidence side, to reduce the refractive index difference on the back surface. Avoiding or reducing the refractive index difference leads to avoiding or minimizing surface damage. 2) Wetting the front surface of the substrate, i.e., the side facing the laser incidence, to reduce contamination on the top surface. Contamination can lead to undesirable damage and / or affect filament formation, meaning the filament may form to a shallow depth. 3) Wetting both sides to reduce the refractive index difference on the back side and to reduce contamination on the top surface of the substrate.

[0105] In this case, the formation of filamentous damaged areas 32,132 under liquid conditions has another advantage: the open hollow chambers are filled with liquid rather than air. This accelerates the subsequent etching process. A special embodiment is the formation of filamentous damaged areas under etching solution as described above.

[0106] Figures 11a, 11b, and 11c each show a perspective view of a substrate, in this example a glass element 200, having a surface structure in which a cavity 5 is formed by the method according to the present invention described above. Figures 11b and 11c show a layered structure comprising two or three glass elements 200. Such structured glass elements 200 (hereinafter also referred to as wafers) and layered structures comprising two or more wafers formed from these glass elements 200 are typically used in the field of microfluidics as so-called microfluidic chips or microfluidic cells. Microfluidic chips or microfluidic cells enable systematic, automated investigation of the chemical or biological properties and reactions of materials. The reactants involved are typically introduced into the flow passage (fabricated groove) and reaction chamber through through-holes 50 in the coverslip, and the reaction products are then analyzed. The method according to the present invention for forming the cavity 5 can be used in the most advantageous form for forming this extremely fine reaction chamber. In other words, the present invention relates to a microfluidic cell 12 comprising a substrate in the form of a glass element having at least one cavity 5.

[0107] Figure 12 shows a top view of a portion of the glass element. A cavity 5 is formed in this glass element by filament formation and etching according to the present invention.

[0108] In this case, the desired region of cavity 5 is >20 μm = d x and 5μm=d yA mesh is formed of filamentous damaged portions 32 with a pitch interval. In this case, the filamentous damaged portions 32 are not formed entirely through the glass volume, but simply formed from one side to the desired depth. During the subsequent etching process, these individual filamentous damaged portions 32 are joined together to form a consistent surface. This forms the cavity 5 according to the present invention. In the illustrated example, the bottom surface of the cavity 5 has a structure with spherical recesses, similar to the side walls.

[0109] Figures 13a and 13b schematically illustrate, in a simplified example for clarity, another arrangement of two wafers 9 and 10, made of glass in this example, as two substrates for use as a microfluidic cell 12. In this case, wafer 9 is flat and planar, while wafer 10, also flat and planar, has two through-holes 122 and a cavity 121. Figure 13b shows the layered structure of the microfluidic cell 12. In this case, the cavity 121 formed by the present invention constitutes a reaction chamber. Both through-holes 122 serve for filling and discharging.

[0110] Therefore, the method according to the present invention for forming filaments on a substrate made of a brittle material, preferably glass or glass ceramics, can be used in particular to form fine and delicate surface structures.

[0111] Therefore, it is particularly well suited for supporting the manufacture of microfluidic chips or microfluidic cells.

[0112] The manufacture of a microfluidic cell according to another aspect of the present invention is described in detail below. The simplest configuration of a microfluidic cell is obtained by combining a lower portion structured by passages with a cover having access to the passages. In the latest prior art, microfluidic cells are manufactured from polymers, for example by injection molding. Corresponding assemblies are known, for example, from European Patent No. 2719460 (EP2719460B1) and German Patent Application Publication No. 102011085371 (DE102011085371A1). However, the manufacture of a microfluidic cell from two polymer components results in the following disadvantages. - The polymer is often incompatible with the solvents used or undergoes unspecified reactions with the biomolecules used (lack of biocompatibility). - The polymer's autofluorescence and limited transparency affect or interfere with the sorting quality when detecting fluorescent marker materials. Furthermore, the polymer surface can only be given functionality by biomarkers in a limited way.

[0113] As a solution to this problem, the manufacture of a microfluidic cell consisting of three components has already been proposed. In this microfluidic cell, the lower and upper components are made of glass, thereby enabling a wide range of functionalities. Furthermore, the passage structure is formed via an organic polymer component or a silicon component. This component is joined to the upper and lower parts by an adhesive that has already been applied before structuring. The polymer component is described in European Patent No. 2547618 (EP2547618B1), and the silicon element is described in Japanese Patent Application Publication No. 2013188677 (JP2013188677A2) and Chinese Patent No. 103992948 (CN103992948B). From European Patent Application Publication No. 3037826 (EP3037826A1), a microfluidic cell having a sandwich structure consisting of an elastomer layer between two glass substrates is also known. Bonding is achieved, for example, by direct bonding of surfaces activated by corona discharge. European Patent Application Publication No. 3088076 (EP3088076A1) also describes a multilayer cell in which a passage structure is fabricated within a silicon layer.

[0114] However, combinations of glass and polymer materials have the drawback that the differing expansion coefficients of the constituent components during analysis under various temperature cycles lead to cell deformation and, in extreme cases, cell non-sealing. Furthermore, this method does not solve the problems of biocompatibility and autofluorescence. In addition, with plastic interlayers, the insufficient strength of the plastic often results in problems where extremely thin and long passage structures cannot be adequately adjusted to the structure of the lower and upper parts during bonding. This adjustment problem becomes even more pronounced because inexpensive manufacturing is only possible by producing large substrates with multiple cells simultaneously.

[0115] A solution to the above requirements is provided, in one aspect of the present invention, by manufacturing the microfluidic cell from three glass components. In these glass components, the intermediate component (also called an insert) consists of a thin, structured glass, which is bonded to the cover and bottom by adhesive applied to both sides after structuring. This ensures that all three components are made of inert, non-autofluorescent, and well-functional materials. No stress is generated due to different high thermal expansions. The use of adhesive technology easily ensures the airtightness of the cell. By applying adhesive to the structured components, the cell-facing surfaces of the bottom and cover can be equipped with biomarkers inexpensively, individually, and comprehensively before the components are bonded. Furthermore, the adhesive technology allows for the absence of encapsulation of particles from plastic, and therefore no longer interferes with the bonding process, while the airtightness of the cell remains guaranteed. Consequently, the requirements regarding the cleanliness of the process environment are correspondingly lower. At least one of the glass components also has a cavity 5 which can be manufactured by the laser-assisted etching method described herein.

[0116] In particular, to manufacture a microfluidic cell, the cavity is processed into a disc-shaped glass element 200, but in this case, - The laser beam 27 of the ultrashort pulse laser 30 is directed towards one of the sides 2 and 3 of the glass element 200, and is focused by the focusing optical element 23 to form an extended focal point on the substrate 1. -The incident energy of the laser beam 27 causes a filamentous damage portion 32 to form in the volume of the glass element 200, and this filamentous damage portion 32 penetrates the volume to a predetermined depth but does not penetrate the glass element 200. -In order to form filamentous damage portions 32, the ultrashort pulse laser 30 is incident with pulse packets comprising one pulse or at least two consecutive laser pulses, and after processing at least two adjacent filamentous damage portions 32, - The glass element is exposed to the etching medium 33, and the etching medium 33 is, - Since at least two filamentous damaged areas 32 are expanded, a cavity is formed, -In short, the structuring method according to the present invention is used for glass elements, - The glass element 200 is bonded to at least one other glass portion on at least one of its two sides, so that the opening of the cavity 5 is closed by the glass portion and a hollow chamber particularly suitable for guiding liquids is formed, and the bond between the glass element 200 and the glass portion is made by an adhesive that is applied, and the opening of the cavity 5 within the glass element 200 is avoided when the adhesive is applied. The adhesive is preferably applied to the side of the glass element 200. However, structured application to the corresponding surface of the glass portion is also possible.

[0117] This yields a microfluidic cell 12 having a disc-shaped glass element 200 having at least one cavity 5 open on one side, and the side of the glass element 200 on the side where the cavity 5 is open is bonded to a glass portion, so that the cavity 5 is closed by the glass portion, forming a hollow chamber suitable for guiding a liquid, sealed between the glass element 200 having the cavity 5 and the glass portion. In this case, the glass element 200 having the cavity 5 and the glass portion are bonded by an adhesive layer, and the adhesive layer has a region left open around the opening of the cavity, so that the portion of the wall of the hollow chamber formed by the glass portion does not have adhesive and is preferably formed by the material of the glass portion that closes the cavity 5.

[0118] According to the advantageous configuration, the microfluidic cell 12 is manufactured by joining at least three glass portions. Thus, a hollow structure suitable for guiding liquids can also be formed by the intermediate glass element 3 having one or more through-openings. These through-openings are then closed when the two other glass portions are joined. Accordingly, it is specified that a disc-shaped glass element having a thickness of up to 700 micrometers, preferably up to 500 micrometers, be structured as follows: The glass element is structured such that it has at least one opening, and this opening connects two parallel sides of the glass element located on opposite sides of the glass element, and each side of the glass element is joined to a glass portion, thereby forming a microfluidic cell having a hollow chamber suitable for guiding liquids, which is sealed between these other glass portions when the opening is closed by the two glass portions. In this case, the joining of the glass element to at least one of the two glass portions is done by an adhesive, and at the time of adhesive application, at least one opening provided in the glass element is avoided. The adhesive is more preferably applied to the side surface of the glass element. However, structured application to the corresponding surface of the glass portion is also possible. At least one of the glass elements, i.e., the intermediate glass element or the outer glass element, has a cavity according to the present invention. The opening of this cavity is closed upon bonding. The cavity may form a separate hollow chamber or be part of a hollow chamber formed by an opening formed in the intermediate glass element.

[0119] Preferably, even thinner glass is used for the intermediate glass element, i.e., glass with a thickness of up to 300 micrometers, for example, 210 micrometers or less. Alternatively, glass elements with a thickness of 100 micrometers or less, for example, up to 70 μm, can be structured and used as glass elements for microfluidic cells. For particularly small structures, thin glass of up to 70 μm, preferably up to 50 μm or even up to 30 μm, can be structured in the opening. In general, the same method as for manufacturing a cavity that is open on only one side, as already described above with respect to Figure 10, is suitable for manufacturing the opening. Accordingly, in one embodiment of the present invention, a through-opening is formed in the substrate 1 in addition to the cavity 5 that is open on one side, particularly by processing a filamentous damaged portion 32 that extends through the substrate and expanding the filamentous damaged portion 32 with an etching medium. Thus, the wall portion of such a through-opening typically has a spherical recess, similar to the cavity 5 that is open on one side. As shown in the example in Figure 10b, the through-opening may extend from the cavity to the opposite side of the substrate. However, it is also possible to form one additional through-opening besides the cavity 5.

[0120] In this regard, Figures 14 and 15 show a cross-sectional view of an example of a microfluidic cell 12 comprising three glass components, namely a glass element 200 and glass components 201 and 202. The glass element 200 has one or more through-openings 101. These through-openings 101 are closed by coupling with the glass components 201 and 202, forming one or more hollow chambers 91 of the microfluidic cell 12 to contain and / or guide liquid in this form. The glass component 201 has a through-opening 401. This through-opening 401 is connected to the hollow chamber 91 and acts as a filling opening in this form.

[0121] The glass portion 201 in the example shown in Figure 14 has a cavity 5 manufactured according to the present invention. Since this cavity 5 is positioned to connect both hollow chambers 91, both hollow chambers 9 are connected through the cavity 5.

[0122] Figure 15 shows an embodiment in which a cavity 5 is machined into a glass element 200 positioned between the two glass portions 201, 202. The cavity 5 forms a passage for connecting, for example, a filling opening 401 to a hollow chamber 91. Both embodiments underlie embodiments of a method in which the opening 401 is present in or machined into at least one of the glass portions 201, 202. In this case, the glass portions 200, 201 are assembled with the glass element 200 such that the opening 401 forms a connection that guides fluid into the hollow chamber 91 formed upon joining.

[0123] The glass portions of the microfluidic cell 12 are bonded together by an adhesive layer 151, as shown in the figure. The opening 101 and cavity 5 do not have the adhesive layer 151. The adhesive is generally applied so that the edges of the through-opening and cavity do not have adhesive.

[0124] Figure 16 shows an assembly for coating an adhesive layer 151 onto the side surfaces 2 of glass portions 201, 202 or glass element 200. Here, the adhesive is applied by a structured printing method. In this printing method, the adhesive is applied to each side surface 2 of the glass element 200 while selectively avoiding a region 131 extending across the opening of a cavity 5 provided in the glass element 200. According to embodiments of the present invention, a printing apparatus 171 equipped with a print head 181 controlled by a computer 199 is used for this purpose. The print head 181 may be, for example, an inkjet print head that moves over the glass element 200 and discharges the adhesive in droplet form during this movement. The computer 199 controls the print head so that the region 131 where the elongated cavity 5 is located is left open. For example, the print head 181 may move meanderingly over the glass element 200. In this case, the print head 181 is reciprocated on the crossbeam and moved in a lateral motion, or alternatively, the glass element 200 is moved in a row. Generally, without being limited to the specified printing or coating method, an improved embodiment of the present invention specifies that the adhesive is applied such that the area 131 without the adhesive layer 151 is larger than the through-opening or cavity to be avoided, thereby specifying that the edge 161 of the applied adhesive layer 151 is spaced apart from the edge 111 of the opening 101 or cavity 5, and is in particular set back.

[0125] The printing method illustrated in Figure 16 is purely illustrative. Other printing methods include tampon printing, screen printing, stencil printing, roll coating, or roll-to-roll coating, coating, and stamp transfer. For particularly large quantities, printing methods such as tampon printing or screen printing are suitable. In one embodiment, for the manufacture of microfluidic cells 12, an acrylic adhesive with a viscosity of 9600 mPa·s is applied to both sides of the glass element 200 by screen printing. Positioning marks allow not only the glass element 200 to be positioned relative to the glass parts 201 and 202, but also the screen printing mask to be positioned relative to the structure of the microfluidic cell.

[0126] The viscosity of the adhesive should generally be suitable for the printing method. Therefore, a low viscosity, for example in the range of 300 mPa·s, is advantageous for tampon printing. In the exemplary inkjet method illustrated in Figure 16, an even lower viscosity, preferably below 50 mPa·s, is advantageous.

[0127] A preferred embodiment of the present invention specifies that a photocurable, preferably UV-curable, adhesive 12 is applied. Therefore, the adhesive 12 can be irradiated with light, preferably UV light, through one of the glass portions 5, 7, so that when the adhesive hardens and is applied to one glass portion or both, both glass portions 201, 202 are firmly bonded to the glass element 3. Suitable adhesives that may be UV-curable may be silicone-containing adhesives, epoxy resins, and acrylates.

Claims

1. A substrate (1) made of a brittle material, preferably a substrate comprising glass or glass ceramics, the substrate (1) having a cavity (5) on at least one side, the walls (8) of the cavity (5) having adjacent chamfered, spherical recesses (7) and the bottom surface of the cavity (5) having the spherical recess (7), the cavity (5) having a depth of up to 5 mm and / or extending into the substrate (1) to a depth of at least 50 μm, preferably at least 100 μm, particularly preferably at least 200 μm, and the taper angle of the walls of the cavity (5) relative to the adjacent side of the substrate (1) is in the range of 90° + / - 5°, preferably 90° + / - 3°, particularly preferably 90° + / - 1°.

2. 2. The substrate (1) according to claim 1, wherein the cavity (5) has at least two regions with different depths.

3. 3. The substrate (1) according to claim 1 or 2, wherein at least one through opening is provided with a wall that is provided with a spherical recess.

4. A substrate (1) described in any one of claims 1 to 3, wherein the cavities (5) have different depths.

5. A substrate (1) described in any one of claims 1 to 4, wherein the cavity (5) has a step portion.

6. A substrate (1) as described in claim 5, wherein the cavity (5) has a step portion in combination with a through opening.

7. A substrate (1) as described in any one of claims 1 to 6, wherein adjacent spherical recesses (7) have mutually adjoining concave rounded portions between them.

8. The substrate (1) of claim 7, wherein the concave rounded portion forms a ridge line.